Evidence is mounting that tumor growth is driven by cancer stem cells (CSCs).These cells are believed to act as the driving force for tumor growth because they are resistant to therapy and are able to self renew. Prior to the discovery of CSCs, the clonal model of tumor progression held the majority of support. This model posits that every cell within a tumor has the capacity for self-renewal and therefore, if not killed during therapy, can repopulate the tumor with subclones resistance to the therapy. The CSC model maintains that a tumor consists of a hierarchy of cells. The CSCs within the tumor are the cells that are able to regenerate themselves and co-produce non-CSC progeny which make up a good deal of the remaining tumor bulk. Because CSCs are a sub population within a group of tumor cells, they are marked with two different molecules that are attached to the outside of the cell. We propose an imaging system based on antibodies that have high affinity for two of these markers which we will develop. We will also develop a third antibody for a small reporting molecule that will be labeled with a dye. This antibody will be divided in half and each half will be fused to the antibodies for the two cell surface markers. When the cell surface markers are bound by the antibodies the two halves will be brought in close proximity and form a binding pocket for the small reporting molecule. We will prove this system works by visualizing whether the small reporting molecule binds by viewing the process on human tumor cells in a microscope. The small reporting molecule will bear a dye that will enable its visualization. The process is outlined below in Aims 1-3. Aim 1. Prepare and test antibodies to the cell surface markers and the small reporting molecule. Aim 2 Fuse the half-antibodies to the whole antibodies and retest to show that binding ability has not been hampered. Aim 3 Add the fused antibodies and dye-labeled small reporting molecule to the human cancer cells and view the process under the microscope to determine whether the system can work. If we are able to make this imaging system work, it can then be tested in animal models of human cancer. If successful, it also be used in humans to view the levels of CSCs during and after therapy. This will be particularly useful when anti-CSC therapies are developed.